Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Polarimetry with the Southern African Large Telescope
David Buckley
SALT Science Director
In collaboration with:Janus Brink (SAAO)Stephen Potter (SAAO)Encarni Romero-Colmenero (SAAO)Ken Nordsieck (Uni. Wisconsin)
COST Action Polarimetry Calibration Workshop: 24 Ja n 2013, Zurich ETH
BASIC ATTRIBUTES
• 10m x 11m PRIMARY MIRROR ARRAY– Spherical Figure (correct spherical aberration at
prime focus)– 91 identical hexagonal 1.2-m segments– Unphased (i.e. not diffraction limited 10-m, just 1-
m) – Mirrors ( Sitall: low expansion ceramic) supported
on a steel structure
• TELESCOPE TILTED AT FIXED 37o
SALT: A Tilted Arecibo-like Optical-IR Telescope mo delled on the Hobby-Eberly Telescope (HET)
• TELESCOPE TILTED AT FIXED 37o
– Declination Coverage +10o
< δδδδ < -75o
– Azimuth rotation for pointing only
• OBJECTS TRACKED OVER 12o
FOCAL SURFACE
– Tracker executes all precision motions (6 d.o.f.)– Tracker contains S pherical A berration C orrector
(SAC) with 8 arcminute FoV ( Prime Focus)– Track objects for ~1 – 3 hours duration
• IMAGE QUALITY– Telescope error budget of ~0.7 arc-second FWHM– Designed to be seeing limited (median = 1 arcsec)
SALT Spherical Aberration Corrector
• Contracted to SAGEM/REOSC (France)• All mirrors coated with LLNL multilayer coating
(Ag/Al)
Mirror M5
60
65
70
75
80
85
90
95
100
Re
flect
ance
Effi
cie
ncy
(%)
Al primary
SAC built
Total
SAC spec
M3 (general asphere) M2M5
M4(convex)
Exit pupil
Focal plane
1.8 metres
50
55
300 800 1300 1800 2300
Wavelength (nm)
Total Spec
SALT/HET Tracking Principle
Tracker off-centreand pupil partially on primary mirror array. At worst extreme, still a ~7 metre telescope.
With tracker and 11-m pupil centred on primary mirror array and central obstruction (from SAC optics), equivalent
Peculiarities of SALT
obstruction (from SAC optics), equivalent to a 9 metre telescope.
Pupil is always underfilledPupil is baffled at exit pupil• controls stray light • used to simulate pupil for calibrations
SALT tracking characteristics
Equivalent to unobscurred9.2-m diameter mirror
Equivalent to unobscurred 7.9-m mirror
Annulus of visibility for SALT:
Annulus represents12.5% of visible sky
Declination range:+10º to -75º
How SALT Observes: Restricted Viewing Window
Observation time available = time taken to cross annulus
But tracker only has limited range ⇒⇒⇒⇒
Additional azimuth moves needed to achieve full obs. time
Implies that all SALT observations have to be queue-scheduled
SALT’s Current Science Instruments
First Generation Instruments chosen to give SALT a wide range of capabilities in UV-VIS range (320 – 900 nm)
• Ensure competitiveness with niche operational modes– UV, Fabry-Perot, high-speed, polarimetry, precision RV
• Take advantage of SALT design and modus operandii– 100% queue scheduled telescope– Capability to react quickly to events, but restricted viewing window
• First two completed & installed from 2005 (“First L ight” instruments)– SALTICAM: a UV-VIS sensitive “video camera” (up to ~1 5 Hz)– Robert Stobie Spectrograph (RSS): a UV-VIS versatile imaging
spectrograph
• Third is the fibre-fed High Resolution Spectrograph (HRS)– Design completed 2005 by UC. Construction by Durham University began in 2007– Commissioning due to begin April 20
The Robert Stobie Spectrograph (RSS)(built at Wisconsin, Rutgers & SAAO)
An efficient and versatile Imaging Spectrograph• capable of UV-Vis spectroscopy from 310 – 900nm using VPHGs (red extension to 1.7µm, using a dichroic,is under construction)• high time resolution ablility (~0.1 s)• specto- and imaging polarimetric capability• Fabry Perot imaging (incl. with pol.)• multiple object spectroscopy
- Can observe ~100 objects at once
Named in memory of Bob Stobie, previous SAAO Director & one the instigators of SALT.
- Can observe ~100 objects at once
RSS reinstalled on SALT (Apr 2011)
RSS Mechanisms
2 Fabry-Perot Etalons (slides)
Polarizing Beamsplitter
(slide)
6-Grating Magazine
40-Slitmask Magazine
3 CCD
2 Polarimeter Waveplates (slides)
Camera Articulation 0 - 100°
Shutter
20-Filter Magazine
3 CCD Mosaic Detector
RSS Polarimetric Science Drivers
Multiple science drivers for RSS polarimetric modes (imaging and spectropolarimetry):
• Polarimetry of “point sources”:– Stars (CVs, Be disks, symbiotics, etc) – AGN, blazars– Supernovae & GRBs– Asteroids– Asteroids
• Polarimetry of “extended sources”:– Galaxies– Reflection nebulae
• Variability– including High Time Resolution (~seconds)– Phase resolved (binaries)
Determining Magnetic Field Strength & Geometry in P olars:Fitting cyclotron model fits to All-Stokes broadban d polarimetry
Single pole system
Example: V834 Cen, P orb= 101 min SAAO 1.9-m photopolarimetry
Fit cyclotron parameters (plasma temp & density, cyclotron opacity, B & θθθθ)
• using Potter’s Stokes imaging technique
• fits model to data using a geneticalgorithm
Extend to spectropolarimetry
Spectropolarimetry with SALT
Spectropolarimetric possibilities:
Time resolved, all-Stokes mode (simultaneous circular + linear)Faint objects (polars + intermediate polars)
e.g. MN Hya: a ~3.4h Polar
Circ. Pol.
Intensity
Cyclotron emission harmonics
RSS Polarimetry
• Uses ½ and ¼ λλλλ Pancharatnam superachromatic waveplates– ½ is made of a 4 x 4 mosaic of 52mm wavelplates wit h effective
102mm diameter– ¼ is a single 60mm waveplate
• 3 x 3 mosaic of calcite beamsplitters– High birefringence separates O & E beams completely on detector– Can’t get single large enough calcite crystals– Mosaic keeps beamsplitter thin, so improves throughput and decreases need – Mosaic keeps beamsplitter thin, so improves throughput and decreases need
for collimated space
½ λWaveplate Mosaic beamsplitter
¼ λWaveplate
Waveplate Configurations for Polarimetry Modes
Waveplates Rotated to specific positions
Beamsplitter splits 2 polarizations
RSS Spectro-/Imaging Polarimetry mode
Waveplate rotates polarization (e.g. 4 or 8 angles)
Polarization of spectra
Also works in Imaging/ Fabry-Perot Modes
RSS Polarimetry
• Imaging polarimetry
CCD 2
CCD 3
CCD 1
FT Boundary
RSS High Speed modes:up to 10 Hz
Allows for:• Fast spectroscopy• Fast narrow band imaging• imaging polarimetry
Configuration for imaging polarimetry
2048 2048 2048
O-rayimage
Focal planeMask
4098
spa
tial
row
s =
8’
4’ x 8’ FoV
image
E-rayimage
4’
4098
spa
tial
row
s =
8’
Configuration for high time speed & time averagedAll-Stokes mode spectropolarimetry
2048 2048 2048
O-rayimage
Focal plane slitmask
Spec.#
1
2
3
4
5
6
Rotate waveplates &Shuffle CCD rows down orbackwards/forwards
Waveplate angles
Spec ½ ¼
E-rayimage
6
7
8
Spec ½ ¼
1 0 0
2 22.5 33.75
3 45 67.5
4 67.5 101.25
5 90 135
6 112.5 168.75
7 135 202.5
8 157.5 236.25Modification of Serkowski’s “all-Stokes” mode, with ¼ waveplateRotating at 1.5 frequency of ½ waveplate, both in s ame direction.
Spectropolarimetry
• O & E beams displaces spatially perpendicular to di spersion
High Speed High Speed SpectropolarimetrySpectropolarimetry
E-ray spectrumO-ray spectrum
Frame transfer boundary
Rotated beamsplitter slot mode spectro-polarimetry
Other Relevant SALT Polarimetric Capabilities(for the study of SNe, mass accreting binaries & SN R)
Some unusual/unique modes:• Low resolution “objective prism” style
spectropol imaging (R ~ 50)– Survey mode for strongly pol. objects?
• Fabry-Perot imaging spectropolarimetry
Polarimetry Commissioning
• Began in 2006• Observations of polarimetric standards (unpolarized & linearly
polarized)• Hampered by optical problems
– Instrument throughput issues (particularly <400 nm)– Appearance of bubbles in Wollaston beamsplitter mosa ic
• Instrument removed Nov 2006 for optical repair (flu id coupling • Instrument removed Nov 2006 for optical repair (flu id coupling issues; A/R coating of one element; sealing of beam splittermosaic)
– Returned in 2010; recommissioning started April 201 1
• Polarimetric recommissioning began in Nov 2011– Curtailed after discovery of catastrophic fluid los s & elastomer de-
bonding– Beamsplitter returned to supplier for refurbishment– New design replaces fluid coupling with pliant glue ing
• Expect beamsplitter retrurn in mid-2013 & completion of polarimetric commissioning
Instrumental Polarization
• Predicted instrumental polarizatised on as function of tracker/pupil geometry
• Based on optical design (segmented M1 & 4 mirrors in aberration corrector)
• Uses Al coatings on hexagonal segments & multi -layer LLNL coatings (more like & multi -layer LLNL coatings (more like dielectric than metal)
Polarimetry Calibration Observations
Standards observed:• Unpolarized
• Polarized
Polarimetry Commissioning
• Typical tracker trajectories
Polarimetry Commissioning
• Results
Polarimetry Commissioning
• Results: unpolarized (HD14069); PG300 grating
Polarimetry Commissioning Results
• Results: polarized (Vela1 #95); PG900
• Results: polarized (HD73882); PG900
Polarimetry Commissioning Results
• Results: polarized (HD73882); PG900
Fringing & Stability Issues
• Fringing seen in derived Stokes parameters
• Due to bubbles in beamsplitter causing interference, which varies over time (e.g. residual for same geometry)
Promise of the future:Example of spectropolarimetric observations of T Pyx
outburst during Apr 2011 recommissioning
• lines all show depolarization => intrinsic polarization (magnitude 0.6%). Likely electron scattering polarization in distorted Nova atmosphere (like in supernova polarization).
SUMMARY & STATUS
• SALT RSS has UV-Vis range of polarimetric capabiliti es
• Will be used to observe a variety of objects
• Commissioning began in 2006, but curtailed due to o ptical issues
• Did manage to characterize instrumental polarizatio n (0.17 – 0.3%)
• Problems of “ripples” in raw O & E spectra due to b ubbles in beamsplitter
• Now awaiting repair and re-commissioning (mid-2013)